INDEX

Explanations

names of specific individuals or characters

oai_token-act-pair · gpt-3.5-turbo

names of individuals or characters

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 archived

-0.74

 infographic

-0.73

 HuffPost

-0.71

 Arbit

-0.71

IDE

-0.67

âĶĢâĶĢ

-0.66

Disclaimer

-0.66

 FANTASY

-0.66

 circular

-0.66

 Thrones

-0.65

POSITIVE LOGITS

ley

1.08

cock

1.08

verson

1.05

cko

1.05

anski

1.05

eman

1.05

inton

1.04

quist

1.03

anson

1.03

enson

1.02

Activations Density 0.256%

names of specific individuals or characters

names of individuals or characters

No Comments

No Known Activations